Instructions to use AiArtLab/sdxs-1b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use AiArtLab/sdxs-1b with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("AiArtLab/sdxs-1b", dtype=torch.bfloat16, device_map="cuda") prompt = "sdxs-1b" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
newvae
Browse files
README.md
CHANGED
|
@@ -95,14 +95,14 @@ FLUX.2-klein-9B | MSE=3.638e-04 PSNR=36.60 LPIPS=0.025
|
|
| 95 |
sdxs-1b | MSE=3.530e-04 PSNR=36.92 LPIPS=0.027
|
| 96 |
|
| 97 |
=== Percents ===
|
| 98 |
-
|
|
| 99 |
-
|----------------------------|-----------|-----------|-----------|-----------|
|
| 100 |
-
| Wan2.2-T2V-A14B | 100% | 100% | 100% | 100% |
|
| 101 |
-
| FLUX.1-schnell VAE | 148.0% | 104.2% | 136.7% | 128.8% |
|
| 102 |
-
| FLUX.2-klein-9B | 244.8% | 110.2% | 209.7% | 165.9% |
|
| 103 |
-
| sdxs-1b | 252.4% | 111.1% | 193.1% | 172.1% |
|
| 104 |
```
|
| 105 |
-
To the best of our knowledge, this VAE achieves SOTA: [eval.py](src/eval.py)
|
| 106 |
|
| 107 |
### Image upscale (vae)
|
| 108 |
```
|
|
|
|
| 95 |
sdxs-1b | MSE=3.530e-04 PSNR=36.92 LPIPS=0.027
|
| 96 |
|
| 97 |
=== Percents ===
|
| 98 |
+
| Model | MSE | PSNR | LPIPS | Edge |
|
| 99 |
+
|----------------------------|-----------|-----------|-----------|-----------|
|
| 100 |
+
| Wan2.2-T2V-A14B | 100% | 100% | 100% | 100% |
|
| 101 |
+
| FLUX.1-schnell VAE | 148.0% | 104.2% | 136.7% | 128.8% |
|
| 102 |
+
| FLUX.2-klein-9B | 244.8% | 110.2% | 209.7% | 165.9% |
|
| 103 |
+
| sdxs-1b | 252.4% | 111.1% | 193.1% | 172.1% |
|
| 104 |
```
|
| 105 |
+
To the best of our knowledge, this VAE probably achieves SOTA: [eval.py](src/eval.py)
|
| 106 |
|
| 107 |
### Image upscale (vae)
|
| 108 |
```
|